R Notebook

knitr::opts_chunk$set(warning = FALSE, message = FALSE)
library(tidyverse)
library(lsa)
library(readxl)
library(plotly)

df <- read_excel('../../data/country_portfolios_WOS.xlsx')

countries_keep <- df %>% group_by(eregroupement) %>% 
  summarise(N=sum(N)) %>% 
  filter(N>1000) %>% 
  pull(eregroupement)

df <- df %>% 
  filter(eregroupement %in% countries_keep)

df_clean <- df %>% 
  select(-Ediscipline) %>% 
  group_by(Type,Especialite,eregroupement) %>% 
  summarise(N = sum(N)) %>% 
  ungroup() %>% 
  mutate(Especialite = factor(Especialite),
         eregroupement = factor(eregroupement)) %>% 
     complete(Type,Especialite,eregroupement,fill=list(N=0))

#filter empty topics

empty_topics <- df_clean %>% 
  group_by(Especialite,eregroupement) %>% 
  summarise(N= sum(N)) %>% 
  filter(N==0)

df_clean2 <- df_clean %>% 
  anti_join(empty_topics)

df_clean <- df_clean %>% 
  group_by(Type,eregroupement) %>% 
  reframe(Especialite,
          N,
          p = N/sum(N))

df_clean2 <- df_clean2 %>% 
  group_by(Type,eregroupement) %>% 
  reframe(Especialite,
          N,
          p = N/sum(N))

cossims <- function(df){
  df %>% 
  select(-eregroupement,-N) %>% 
  pivot_wider(names_from = Type,values_from = p) %>% 
  summarise(cos_IM = cosine(International,Mobile),
            cos_IN = cosine(International,National),
            cos_MN = cosine(Mobile,National))
}

cosine_sim <- df_clean %>% 
  group_by(eregroupement) %>% 
  cossims()

size versus change

ggplotly(
df %>% group_by(eregroupement) %>% 
  summarise(N=sum(N)) %>% 
  right_join(cosine_sim) %>% 
  pivot_longer(cols = cos_IM:cos_MN,names_to = 'relation',values_to = 'cosine',names_prefix = 'cos_') %>% 
  mutate(relation = case_when(relation=='IM' ~'International-Mobile',
                              relation=='IN' ~'International-National',
                              relation=='MN' ~'Mobile-National',
                              )) %>% 
  filter(eregroupement!='ZZALL') %>% 
  ggplot(aes(N, cosine, color=relation, label=eregroupement)) +
  geom_point()+
  scale_x_log10()
)

Try to normalize size

compare all papers vs all papers minus (international/Mobile/National)

cossims_diff <- function(df){
  df %>%
  select(-eregroupement,-p) %>% 
  pivot_wider(id_cols = c(eregroupement,Especialite),names_from = Type,values_from = N) %>% 
  mutate(All = International + Mobile + National,
         International = All - International,
         Mobile = All - Mobile,
         National = All - National) %>% 
  mutate(International = International/sum(International),
         Mobile = Mobile/sum(Mobile),
         National = National/sum(National)) %>% 
  summarise(International = cosine(International,All),
            National = cosine(National,All),
            Mobile = cosine(Mobile,All))
}

cosine_sim_diff <- df_clean %>% 
  group_by(eregroupement) %>% 
  cossims_diff()

ggplotly(
df %>% group_by(eregroupement) %>% 
  summarise(N=sum(N)) %>% 
  right_join(cosine_sim_diff) %>% 
  pivot_longer(cols = International:Mobile,names_to = 'relation',values_to = 'cosine') %>% 
  filter(eregroupement!='ZZALL') %>% 
  ggplot(aes(N, cosine, color=relation, label=eregroupement)) +
  geom_point()+
  scale_x_log10()
)

group sizes

ggplotly(
df %>% group_by(eregroupement, Type) %>% 
  summarise(N=sum(N)) %>% 
  right_join(cosine_sim_diff %>% pivot_longer(International:Mobile, names_to='Type', values_to='cosine')) %>% 
  # pivot_longer(cols = International:Mobile,names_to = 'relation',values_to = 'cosine') %>% 
  filter(eregroupement!='ZZALL') %>% 
  group_by(eregroupement) %>% 
  mutate(country_N = sum(N)) %>% 
  ggplot(aes(country_N, cosine, color=Type,size=N, label=eregroupement)) +
  geom_point()+
  scale_x_log10()
)

NA

cosine_sim_diff <- df_clean %>% 
  group_by(eregroupement) %>% 
  cossims_diff()
cosine_sim_diff2 <- df_clean2 %>% 
  group_by(eregroupement) %>% 
  cossims_diff()

cosine_sim_diff %>% select(-eregroupement)-
cosine_sim_diff2 %>% select(-eregroupement)

NA

cosine_sim_diff2 <- df_clean2 %>% 
  group_by(eregroupement) %>% 
  cossims_diff()

ggplotly(
df %>% group_by(eregroupement, Type) %>% 
  summarise(N=sum(N)) %>% 
  right_join(cosine_sim_diff2 %>% pivot_longer(International:Mobile, names_to='Type', values_to='cosine')) %>% 
  # pivot_longer(cols = International:Mobile,names_to = 'relation',values_to = 'cosine') %>% 
  filter(eregroupement!='ZZALL') %>% 
  group_by(eregroupement) %>% 
  mutate(country_N = sum(N)) %>% 
  ggplot(aes(country_N, cosine, color=Type,size=N, label=eregroupement)) +
  geom_point()+
  scale_x_log10()
)

X-axis with the gini of the concentration of topics